ABSTRACT. Beliefs about epistemic justification come into play when people read and try to comprehend controversial scientific issues with social implications, such as climate change, the safety of nuclear power, and whether sun exposure is healthy or harmful. Drawing on theoretical frameworks concerning epistemic cognition, I will summarize research on justification beliefs and comprehension, the relationship between epistemic beliefs and beliefs about the topic itself, and educational implications of changing students’ beliefs about epistemic justification.

ABSTRACT. Scientists’ perceived credibility has a significant impact on trusting scientific knowledge. However, less is known about people’s rationale and thought processes behind their credibility judgments. Therefore, we examined how variations in trustworthiness and expertise influence perceived source credibility of health and climate scientists, and to what extent people’s self-explanations of their credibility evaluations differ by these variations. Overall, people rely on both trustworthiness and expertise to establish credibility, although their relative importance can differ.

ABSTRACT. This study explores the intersection of sourcing abilities and advanced theory of mind (AToM) in evaluating Internet misinformation, emphasizing their critical role in discerning online information credibility. By examining these cognitive skills' impact on critical reading, the research offers insights into the nuanced process of assessing digital content's reliability, highlighting the significance of source evaluation in the digital age.

ABSTRACT. Critique generation, or students’ abilities to elaboratively specify texts’ inaccuracies, omissions, or imprecisions, constitutes a key critical reading outcome and one that needs to be explicitly scaffolded. This study constitutes an initial step in developing an intelligent tutoring system to scaffold critique. We automatically classify students’ responses to texts as critiques or non-critiques, with 75.76% accuracy. Additionally, terms identified as having high feature importance coincide with qualitative critique-type categories identified by human raters.

ABSTRACT. A sample of freshmen undergraduates (n = 127) received an intervention to promote the critical evaluation of Internet sources in one of two versions: asymmetric (expert instructor) or symmetric (peer mentor). Both groups outperformed a control group (no intervention). Additionally, the symmetric modality showed greater gains and more intense epistemic emotions than its asymmetric counterpart.

ABSTRACT. Reward devaluation theory (RDT) posits that depressed individuals exhibit a bias toward negative information, devaluing positive stimuli. This study investigates the manifestation of reward devaluation in a novel, ecologically valid Valence Selection Task. Across three experiments, participants select between positive, negative, or neutral sentence endings. Depressive symptoms correlate with avoidance of positive responses, supporting RDT. These findings underscore the relevance of RDT and its implications for understanding interactions between textual valence and reward devaluation.

ABSTRACT. This study examines the extent to which 101 ​​readers’ judgments of the stance expressed in six tweets about Roe v. Wade align with posters’ reports of what they intended the tweets to express. Readers’ alignment with the posters’ self-reported stance in each tweet was surprisingly low, ranged substantially across readers, and was not predicted by factors like Twitter experience. Undetected conceptual misalignment may be an underappreciated factor in social media communication.

ABSTRACT. People read brief passages and made memorial judgements about explicit, paraphrased, inferential, and lure test probes. Different groups based their judgments either on verbatim, gist, or implicational criteria. Multinomial tree processing analysis indicated that the illusory or phantom recollection of vivid discourse details influenced judgements about paraphrases and confirmed phantom recollective support for bridging inferences but not elaborative inferences. The results clarify the representational profile of these probe types across surface, textbase, and situational representations.

ABSTRACT. Solving a problem with an Aha! impacts memory (the Insight memory advantage). Auble et al.(1979) suggested a similar effect using initially incomprehensible sentences. They found better recall for sentences that became comprehensible after cues were presented, and attributed this to Aha! experiences. However, they had no measure of Aha! experiences. This study adds a measure of Aha! experiences to provide direct evidence that not only comprehension, but the way that understanding emerges, affects memory.

ABSTRACT. Both self-explanation (during reading) and explanatory retrieval practice (after reading) can support comprehension. This study used a 2(online: think-aloud, self-explain) x 3(offline: reread, recall, explanatory retrieval) design to examine how combinations of these strategies affect science text comprehension after a one-week delay. Analyses demonstrated a self-explanation effect, but no retrieval practice effect. This was qualified by an interaction in which the self-explanation effect was strongest when combined with retrieval practice, but not explanatory retrieval.

ABSTRACT. This study examined the interplay of intrinsic reading motivation, reading self-concept and reading task value, as well as task model on 6th graders functional reading performance. Three hundred and forty students completed a reading motivation questionnaire created for the study, four reading comprehension tasks involving multiple documents, and a task model assessment. Reading motivation, especially self-concept, predicted reading comprehension and task model. Most importantly, task model mediated the relation between reading motivation and reading comprehension.

ABSTRACT. Answering comprehension questions requires test takers to understand what the question asks and what process they need to engage in to find an answer. This study attempted to train undergraduate students to recognize whether questions can be answered literally or whether it requires an inference, and explored the impact on performance on PISA-like comprehension tests. Performance on the categorization task predicted performance on the comprehension test. Categorization practice with feedback improved comprehension scores.

ABSTRACT. The present study examined the effect of generating one’s own drawings versus viewing provided drawings on the accuracy of students’ judgments of learning (JOL) immediately and after a delay. Overall, there was no effect of generating drawings versus viewing provided drawings on comprehension, which contrasts with research demonstrating benefits for learner-generated drawing tasks. Further, generating drawings did not result in improved comprehension monitoring compared to viewing provided drawings.

ABSTRACT. Undergraduate students received reading strategy instruction and opportunities for deliberate practice via the iSTART intelligent tutoring system or not (i.e., no-treatment control group). Participants’ reading comprehension and psychology knowledge were assessed. The iSTART group demonstrated substantially greater scores than the control group on a post-training reading comprehension measure (Cohen’s d > 1.0). The iSTART post-training and control group average psychology knowledge scores did not differ, but overall these scores were unexpectedly low.

ABSTRACT. Twenty-one middle-aged adults (mean age 54 years; 14 males) and twenty-one senior adults (mean age 78 years; 14 males) were administered a narrative production task and tests that explored their attentional and inhibitory control skills. Senior adults had reduced lexical selection and narrative organization skills that correlated with measures of sustained attention and inhibitory control.

ABSTRACT. We propose that taking a literary character’s perspective entails finding analogous (but not necessarily identical) experiences from one’s own life. To test this account, we asked readers, before reading a story, to retrieve either a story-related or story-unrelated prior episode from their life. As predicted, retrieving related (but not unrelated) experiences improved perspective taking, and the strength of the related memory predicted the magnitude of the effect.

ABSTRACT. Past literature suggests that readers prefer that good characters experience good outcomes and bad characters bad outcomes. However, we investigated whether people judge bad characters as deserving certain highly positive person-resilient outcomes related to concepts of human dignity and natural rights (such as surviving cancer). In five experiments, we found that participants judged negative characters as more deserving of person-resilient outcomes compared to more ordinary positive outcomes and highly-positive, but not person-resilient, outcomes.

ABSTRACT. Despite the immense popularity of multi-episode narratives, theoretical understandings of memory for and comprehension of serialized fiction remain understudied in discourse psychology. We examined participants’ emotions and perceived memory for scenes that varied in narrative density from a television series. Higher emotional valence predicted higher memory ratings for high-density events, lower memory ratings for low-density events, and lower ratings of memory for details. Higher arousal predicted higher ratings of overall memory and memory for details.

ABSTRACT. Swearing occurs frequently in daily conversation and is typically used to indicate strong emotions that are not easy to convey with traditional non-taboo words. One particularly strong factor that impacts swearing is gender. The current study investigated how the structure and patterns of taboo-language use differ from typical language use in popular movies, focusing on gender dynamics. Using social network analysis, taboo-interactions were found to involve drastically fewer dialogue exchanges, participating characters, and reciprocation.

ABSTRACT. Task instructions are used to guide students’ reading goals. However, different students may conceptualize the same task instructions differently and develop different reading goals. We investigated the formation of readers’ task models (i.e., reading goals and plans to complete them) across task instructions and examined their effect on reading processes and products. Findings suggest that task instructions may shape the task model, which in turn influences reading processes and products.

ABSTRACT. In this review we evaluated the current research on pre-reading task instructions. We describe the different approaches researchers have used to manipulate pre-reading task instructions and identified how different pre-reading task instructions impacted reading comprehension. The findings revealed that pre-reading task instructions differed with respect to what reading content was signaled, what reading strategies were prompted, the specificity of the reading purpose, and other factors. Each had distinct effects on reading comprehension.

ABSTRACT. Knowledge of words is critical to utilizing reading strategies and comprehending text. The present study utilized NLP tools and constructed responses to explore whether features of the language people used when thinking-aloud to text were related to their vocabulary knowledge and use of reading strategies (bridging, elaboration). Results suggested that indices of lexical sophistication were related to vocabulary scores and that these same indices were differentially related to bridging, elaboration, and reading comprehension scores.

ABSTRACT. Motion verbs like come and go make reference to perspectival representations, but a satisfactory account of their behavior remains elusive. Come describes motion towards the Speaker (or Addressee) and go describes motion away from Speaker/Addressee, However, some claim that perspective shift to a third person is possible. We test this experimentally, and find that deictic perspective-shift is much less available than previously claimed. This challenges prior claims and provides a foundation for future work.

ABSTRACT. This study investigates the impact of a source evaluation training on secondary school students. A pretest-posttest design involving 57 participants showed significant improvements in reading comprehension and critical thinking post-training, with no changes in source memory tasks. Correlations between pretest scores highlight significant associations between source memory and performance in reading comprehension and critical thinking. These findings offer valuable insights for educational strategies.

ABSTRACT. We investigate reading skill profiles of at-risk college students using a multidimensional assessment battery. An unintended administration error led to a subset of students scoring at ceiling level on battery subtests. However, varied patterns emerged among students scoring at ceiling on different subtests. Results challenge assumptions about reading skill components necessary for post-secondary academic success. The study underscores the importance of revisiting assessment practices to provide effective support for diverse post-secondary students.

ABSTRACT. Skilled reading involves the efficient coordination of several print and linguistic skills. Many students at the middle-grades level do not have well-developed foundational reading skill sets, and cannot benefit from comprehension instruction as efficiently as their peers. We examined a large-scale dataset of this population using a reading skills assessment. Students with better foundational skills are more likely to score higher on comprehension measures, and the size of this gap increases across grade levels.

ABSTRACT. Texts’ genre is considered an important aspect of sources. An essential distinction is the one between texts claiming to convey true information (i.e., nonfictional texts) and texts not making this claim (i.e., fictional texts). In order to distinguish between these types of texts, readers need knowledge about textual fictionality. I present a novel test of knowledge about textual fictionality including an IRT analysis as well as tests of assumptions regarding item difficulty and criterion-related validity.

ABSTRACT. This study examined the role of the causal connectivity of the statements, note-taking condition and modality of presentation in the comprehension of materials about HIV Health Education. We asked a group of Chilean college students to listen to or read the material, perform a note-taking task, and then to perform a main-ideas recall task and a speakers’ emotions recall task. Results indicated that establishing meaningful connections interplays with note-taking in written and spoken discourse comprehension.

ABSTRACT. The actual widespread use of digital technologies has created the need to learn to manage simultaneous multiple documents when reading. The goal of this scoping review has been to examine the tasks used to promote and measure the processing of multiple documents, both complementary and conflicting. Findings reveal that instructions encouraging across-text connections are especially beneficial for complementary documents, whereas requests to complete an argumentative task show an advantage for contradictory texts.

ABSTRACT. How does the legal text of international regimes change over time? We use document embeddings to analyze the Texts of Trade Agreements database. We build a system-wide model of linguistic similarity and dissimilarity in the trade regime, allowing a birds-eye-view of the language of the trade regime over time for the first time. We find that similarity between texts in the regime fluctuates over time, revealing periods of convergence and divergence in the trade regime.

ABSTRACT. In two experiments, we investigated whether pictures change the validation process. We used the contradiction paradigm and investigated whether illustrating the text information inconsistent with a later target sentence can increase the awareness of the inconsistency and whether the validation process changes if the inconsistency is evident only between a picture and a later target sentence. Results suggested that pictures did not increase the awareness of inconsistencies and that the validation process was more text-driven.

ABSTRACT. The present study aims to promote the application of two epistemic strategies, sourcing and corroboration, in a four-lesson teacher-led strategy training to 7th- and 8th-graders in eleven classes. Effects are compared to students in ten classes receiving a training on conceptual knowledge about misinformation. Data collection involves a pretest, a posttest (one week after final lesson), and a follow-up-test (4-5 weeks after final lesson), resulting in a quasi-experimental between (training variant)-within (points of measurement)-subjects design.

ABSTRACT. We assess whether the sentiment of one-word descriptions written for satirical news texts predict awareness of satirical intent. We calculate sentiment values from SentiWordNet and use these values to predict perceptions of sincerity, a proxy for recognition of satirical intentions. Results show negative sentiment explained variation in sincerity scores, whereas positive sentiment did not. These results align with theoretical models of satirical discourse and provide another means to assess readers’ ability to infer satirical intent.

ABSTRACT. This study combines cognitive linguistics and dialogic syntax to examine discourse development in naturally emerging interactional dialogue. Cognitive models appear to embrace the dynamic nature of discourse development with intonation units and given/accessible/new information, while it neglects the dialogic features. The dialogic analysis integrates relational and interactional features, revealing negotiated stances and epistemic cognition that are essential to the direction of discourse building. Intersubjectivity is redefined based on the dialogic nature of language.

ABSTRACT. We describe efforts to design and test an online critical thinking module that is aimed to assess and support the development of argument evaluation skills. We collected initial empirical evidence from 84 college students with various backgrounds. Our pilot data suggested that the module may reliably assess students’ argument evaluation skills, and student feedback indicated an overall positive experience. In this presentation, we will discuss the module design, study results, and implications in detail.